A Matrix of Feedback for Learning: A Brief Summary 

In this post, we provide a summary of a recent study that sought to determine which types and levels of feedback are most common in schools in Brisbane, Queensland. The authors of the study (Brooks, Carroll, Gillies, & Hattie, 2019) noted the following rational and justification for their investigation: 

  • The provision of feedback is not a guarantee of learning (feedback is typically viewed as a one-way transformative process with more emphasis placed on the fact that it is given and less attention paid to how it’s being received);
  • Meta-analyses show that there are highly variable effects of feedback (often the detrimental impact of praise is overlooked). 

What Makes Feedback More Effective? 

In reviewing the conditions that make feedback more effective, the authors present Hattie and Timperley’s (2007) model and note that it encompasses each condition. These conditions include: (1) clear expectations, criteria, and standards for the learner at the beginning or during the learning cycle; (2) ongoing and specific feedback delivered during the learning period; (3) feedback that fosters self-regulation; and (4) opportunities for learners to act on earlier feedback that was received or self-generated. Furthermore, the model includes three types of feedback (feed up, feed back, and feed forward) and four levels of feedback (task, process, self-regulation, and self). 

Previous research (Harris, Brown, & Harnett, 2015) used Hattie and Timperley’s (2007) model as a conceptual framework to investigate which level of feedback was directed to during peer and self-assessment and found most feedback was directed to the task level. Harris et al., also concluded that coding feedback using Hattie and Timperley’s model was a suitable form of analysis for the research. Therefore, ultimately, the authors of this study sought to determine how to apply the conceptual model to practical conditions in the classroom. 

What’s Important to Note? 

What is important to note about the feedback types: 

Feedback Type Answers the question… For example, the teacher may say…
Feed up Where am I going?  “I am looking for your ability to persuade your audience.”
Feed back  How am I going?  “You are using persuasive devices effectively to persuade your audience.” 
Feed forward  Where to next?  “You could strengthen your argument by evaluating ideas from the

opposing point of view.”

What is important to note about feedback levels:

Task Level  Process Level  Self-Regulation Level Self-Level
How well tasks are  performed/understood.  The main processes needed to perform/understand the task. Self-monitoring, directing, and regulating actions. Personal evaluations and affect (usually positive) about the learner. 

What is also important to note is that given Hattie and Timperley’s description of self-level feedback as potentially having negative effects upon learning, the researchers omitted self-level feedback level from their analysis in this study. 

What Did the Researchers Find?

Upon transcribing classroom conversations from a voice recorder and using a protocol for coding the data, the researchers found the following: 

  1. Feed back (i.e., How am I going?) was the most common type of feedback provided. 
  2. Feed forward (i.e., Where to next?) was the least common type of feedback provided. 
  3. Task level feedback was 4 x more prevalent than process level feedback. 
  4. Task level feedback was 12 x more prevalent than self-regulatory level feedback. 

When intersecting the feedback type and feedback level, the following was observed: 

  1. Feed up (i.e., Where am I going?) was most commonly directed at the task level (and predominantly directed to the whole class).
  2. Self-regulatory feed up was rarely recorded. 
  3. Feed up at the process level was directed more to individual students and was characterized by the use of prompts and questions. 
  4. Feed back was most commonly directed at the task level. 
  5. Feed forward was most commonly directed at the task level. 
  6. Self-regulatory feed forward was the least likely feedback level observed. 

The Authors’ Conclusions: 

The authors list several implications for classroom practice. 

  1. Feed up is an important process to help clarify for learners the learning intentions and success criteria.
  2. Feed forward, the least frequent feedback type identified, is a vital stage of the feedback process as it closes the gap between where students are now and where they need to be.
  3. Most feedback was directed to the task level which is associated with the promotion of surface learning. 
  4. Relatively little feedback was directed to process and self-regulatory feedback which are more likely to engender deeper and relational learning (Hattie, 2012). In other words, the feedback that is the most transformative for learners was the least prevalent.

The authors noted that feedback directly related to the self regulatory level, although the least frequent feedback recorded in the study, was largely centred on the student goals derived from the success criteria which focused upon the standards required for mastery learning of the subject. This reflection and individualised learning goals at the self regulatory level target the specific needs of the learner and used as guidance for improvement. 

Findings from this study were subsequently used to inform the development of a feedback matrix that bridges research to practice.  This matrix can then be used for professional development purposes to understand the interactive nature of the types and levels of feedback that promote progress in learning. 

A Matrix of Feedback for Learning


Brooks, C.,Carroll, A., Gilles, R.M., &Hattie, J. (2019). A Matrix of Feedback for Learning. Australian Journal of Teacher Education, 44(4) 

Written by

Jenni Donohoo, Sue Bryen, and Brian Weishar are co-authors of “Implementing High-Leverage Influences from the Visible Learning Synthesis: Six Supporting Conditions.”

No comments

leave a comment

8 + three =